An important step in antisense technology is identification of mRNA sites that can be targeted efficiently. Many experiments have shown that certain antisense oligonucleotides are more efficient than others in suppressing specific gene expression. A routine approach to finding the most active antisense oligonucleotides involves the synthesis of numerous oligonucleotides, (often up to several dozen) complementary to different regions of the targeted mRNA, followed by activity screening of the oligonucleotides in cells. Alahari, S. K. et al., Mol. Pharmacol., 1996, 50, 808-19; Bennett, C. F., Condon, T. P., Grimm, S., Chan, H. & Chiang, M. Y., J. Immunol., 1994, 152, 3530-40; Chiang, M. Y. et al., J. Biol. Chem., 1991, 266, 18162-71; Dean, N. M., McKay, R., Condon, T. P. & Bennett, C. F., J. Biol. Chem., 1994, 269, 16416-24; Dean, N. M. et al., Biochem. Soc. Trans., 1996, 24., 623-9; Duff, J. L., Monia, B. P. & Berk, B. C., J. Biol. Chem., 1995, 270, 7161-6; Lee, C. H. et al., Shock, 1995, 4, 1-10; Lefebvre d'Hellencourt, C., Diaw, L., Cornillet, P. & Guenounou, M., Biochim. Biophys. Acta, 1996, 1317, 168-74; Miraglia, L., Geiger, T., Bennett, C. F. & Dean, N. M., Int. J. Immunopharmacol., 1996, 18, 227-40; Stepkowski, S. M., Tu, Y., Condon, T. P. & Bennett, C. F., J. Immunol., 1994, 153, 5336-46; J. Immunol., 1995 154, 1521; Stewart, A. J. et al., Biochem. Pharmacol., 1996, 51, 461-9. Strategies for increasing the percentage of active antisense oligonucleotides should have significant benefit. A variety of such strategies have been tried, using a range of criteria of oligonucleotide design.
The calculated Gibbs free energy (ΔG) values for duplex formation between an oligonucleotide and mRNA molecule correlates with oligonucleotide antisense activity (Stewart, A. J. et al., Biochem. Pharmacol., 1996, 51, 461-469), though hybridization affinity alone is not sufficient to ensure antisense oligonucleotide efficiency in cells. Chiang et al., J. Biol. Chem., 1991, 266, 18162-18171. Systematic alignment of computer-predicted local RNA secondary structures has also been attempted in selecting antisense oligonucleotides for inhibition of intracellular adhesion molecule-1 (ICAM-1) expression. Patzel et al., Nucl. Acids Res., 1999, 27, 4328-4334. The predicted stabilities of antisense oligonucleotide:target-RNA duplexes and their competition with predicted secondary structures of both the targets and antisense oligonucleotides have also been studied as a means to predict antisense efficacy. Stull et al., Nucl. Acids Res., 1992, 20, 3501-3508; Mathews et al., RNA, 1999, 5, 1458-1469. In spite of these proposed methods, however, active antisense compounds are still generally identified through empirical testing of multiple antisense sequences.
Sequence-based strategies for predicting antisense efficacy have also been tried. Smetsers et al. have found that there is a bias in the nucleotide composition of active antisense oligonucleotides, with GG, CCC, CC, GAC and CG motifs significantly overrepresented and TT and TCC significantly underrepresented in a database of 206 antisense oligonucleotides from the published literature (G, C, A and T represent the four major DNA nucleobases guanine, cytosine, adenine and thymine, and/or their corresponding nucleosides guanosine, cytidine, adenosine and thymidine). However, some of these motifs are reported in the literature to induce nonantisense effects such as by protein binding or mitogenic mechanisms. The authors conclude that further analysis is needed to investigate whether these motifs should be avoided or favored in oligonucleotide design. Smetsers et al., Antisense Nucl. Acid Drug Dev., 1996, 6, 63-67.
While Smetsers found that CCC is overrepresented and TCC is underrepresented among active antisense sequences, Tu et al. found that the TCCC motif is overrepresented among the most active oligonucleotides compared to their inactive counterparts. Tu, G. et al., J. Biol. Chem., 1998, 273, 25125-25131 and WO 99/01139. In an analysis of published oligonucleotide sequences and in prospective experiments with TNF-α mRNA, oligonucleotides containing the TCCC motif were found to have a much higher success rate (50%) than oligonucleotides selected by trial and error (6%). Tu, G. et al., J. Biol. Chem., 1998, 273, 25125-25131. Antisense oligonucleotides for inhibiting TNF” which comprise a TCCC motif are claimed. Methods of making antisense oligonucleotides which target a sequence comprising a GGGA motif and methods of inhibiting gene expression with these oligonucleotides are also disclosed and claimed. Methods of predicting the efficacy of an antisense oligonucleotide by determining whether the oligonucleotide is complementary to an RNA sequence containing a GGGA motif are also claimed. WO 99/01139.
Non-antisense biological effects have been observed when cells are treated with phosphorothioate oligonucleotides containing four contiguous guanosines. These oligonucleotides were demonstrated to form tetrameric (four-stranded) structures. Non-antisense effects have also been observed with phosphorothioate and phosphodiester oligonucleotides that contain several sets of consecutive guanosines. The G-quartet structure has been implicated in oligonucleotide binding to proteins. Wyatt, J. R. and Stein, C. A., Appl. Antisense Ther. Restenosis, 1999, pp. 133-140. Kluwer, Boston, Mass. Oligonucleotides containing the sequence GGGG (G4) have been found to have antiviral activity against a number of viruses. Sequences containing a G4 motif or two or more G3 motifs has been found to be effective antivirals. It has also been reported that oligonucleotides containing a conserved G4 core sequence or two stretches of 3 G's are effective inhibitors of phospholipase A2 activity and that such oligonucleotides could be useful for modulation of telomere length on chromosomes. Wyatt et al., Proc. Natl. Acad. Sci. U.S.A., 1994, 91, 1356-1360; Wyatt et al., Biochemistry, 1996, 35, 8002-8008; U.S. Pat. Nos. 5,523,389, 5,756,710 and 5,952,490; WO 94/08053.
According to the present invention, a series of oligonucleotide sequence motifs have been identified which, when present in an antisense oligonucleotide, have been demonstrated to be either positively or negatively correlated with antisense activity of the oligonucleotide. Consequently, a series of target sequence motifs complementary to the oligonucleotide sequence motifs have correspondingly been demonstrated to be either positively or negatively correlated with antisense activity when an oligonucleotide is targeted to a nucleic acid sequence containing such target sequence motifs. Several dozens of such motifs have been identified. Compositions and methods using these motifs are herein provided.